How to understand limitations of generative networks
Well-trained classifiers and their complete weight distributions provide us with a well-motivated and practicable method to test generative networks in particle physics. We illustrate their benefits for distribution-shifted jets, calorimeter showers, and reconstruction-level events. In all cases, th...
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Dokumenttyp: | Article (Journal) |
| Sprache: | Englisch |
| Veröffentlicht: |
25 January 2024
|
| In: |
SciPost physics
Year: 2024, Jahrgang: 16, Heft: 1, Pages: 1-32 |
| ISSN: | 2542-4653 |
| DOI: | 10.21468/SciPostPhys.16.1.031 |
| Online-Zugang: | Verlag, lizenzpflichtig, Volltext: https://doi.org/10.21468/SciPostPhys.16.1.031 Verlag, lizenzpflichtig, Volltext: https://scipost.org/10.21468/SciPostPhys.16.1.031 
|
| Verfasserangaben: | Ranit Das, Luigi Favaro, Theo Heimel, Claudius Krause, Tilman Plehn and David Shih |
MARC
| LEADER | 00000caa a2200000 c 4500 | ||
|---|---|---|---|
| 001 | 1890705632 | ||
| 003 | DE-627 | ||
| 005 | 20240703183559.0 | ||
| 007 | cr uuu---uuuuu | ||
| 008 | 240605s2024 xx |||||o 00| ||eng c | ||
| 024 | 7 | |a 10.21468/SciPostPhys.16.1.031 |2 doi | |
| 035 | |a (DE-627)1890705632 | ||
| 035 | |a (DE-599)KXP1890705632 | ||
| 035 | |a (OCoLC)1443679435 | ||
| 040 | |a DE-627 |b ger |c DE-627 |e rda | ||
| 041 | |a eng | ||
| 084 | |a 29 |2 sdnb | ||
| 100 | 1 | |a Das, Ranit |e VerfasserIn |0 (DE-588)1331721989 |0 (DE-627)1890705993 |4 aut | |
| 245 | 1 | 0 | |a How to understand limitations of generative networks |c Ranit Das, Luigi Favaro, Theo Heimel, Claudius Krause, Tilman Plehn and David Shih |
| 264 | 1 | |c 25 January 2024 | |
| 300 | |b Illustrationen | ||
| 300 | |a 32 | ||
| 336 | |a Text |b txt |2 rdacontent | ||
| 337 | |a Computermedien |b c |2 rdamedia | ||
| 338 | |a Online-Ressource |b cr |2 rdacarrier | ||
| 500 | |a Veröffentlicht: 25. Januar 2024 | ||
| 500 | |a Gesehen am 05.06.2024 | ||
| 520 | |a Well-trained classifiers and their complete weight distributions provide us with a well-motivated and practicable method to test generative networks in particle physics. We illustrate their benefits for distribution-shifted jets, calorimeter showers, and reconstruction-level events. In all cases, the classifier weights make for a powerful test of the generative network, identify potential problems in the density estimation, relate them to the underlying physics, and tie in with a comprehensive precision and uncertainty treatment for generative networks. | ||
| 700 | 1 | |a Favaro, Luigi |d 1996- |e VerfasserIn |0 (DE-588)1331722098 |0 (DE-627)1890706329 |4 aut | |
| 700 | 1 | |a Heimel, Theo |d 1996- |e VerfasserIn |0 (DE-588)1217495584 |0 (DE-627)1730495281 |4 aut | |
| 700 | 1 | |a Krause, Claudius |e VerfasserIn |0 (DE-588)1114848948 |0 (DE-627)869248847 |0 (DE-576)477537456 |4 aut | |
| 700 | 1 | |a Plehn, Tilman |d 1969- |e VerfasserIn |0 (DE-588)1021935573 |0 (DE-627)715839535 |0 (DE-576)363449809 |4 aut | |
| 700 | 1 | |a Shih, David |e VerfasserIn |4 aut | |
| 773 | 0 | 8 | |i Enthalten in |t SciPost physics |d Amsterdam : SciPost Foundation, 2016 |g 16(2024), 1, Artikel-ID 031, Seite 1-32 |h Online-Ressource |w (DE-627)881391751 |w (DE-600)2886659-9 |w (DE-576)484813447 |x 2542-4653 |7 nnas |a How to understand limitations of generative networks |
| 773 | 1 | 8 | |g volume:16 |g year:2024 |g number:1 |g elocationid:031 |g pages:1-32 |g extent:32 |a How to understand limitations of generative networks |
| 856 | 4 | 0 | |u https://doi.org/10.21468/SciPostPhys.16.1.031 |x Verlag |x Resolving-System |z lizenzpflichtig |3 Volltext |
| 856 | 4 | 0 | |u https://scipost.org/10.21468/SciPostPhys.16.1.031 |x Verlag |z lizenzpflichtig |3 Volltext |
| 951 | |a AR | ||
| 992 | |a 20240605 | ||
| 993 | |a Article | ||
| 994 | |a 2024 | ||
| 998 | |g 1021935573 |a Plehn, Tilman |m 1021935573:Plehn, Tilman |d 130000 |d 130300 |e 130000PP1021935573 |e 130300PP1021935573 |k 0/130000/ |k 1/130000/130300/ |p 5 | ||
| 998 | |g 1114848948 |a Krause, Claudius |m 1114848948:Krause, Claudius |p 4 | ||
| 998 | |g 1217495584 |a Heimel, Theo |m 1217495584:Heimel, Theo |d 130000 |d 130300 |e 130000PH1217495584 |e 130300PH1217495584 |k 0/130000/ |k 1/130000/130300/ |p 3 | ||
| 998 | |g 1331722098 |a Favaro, Luigi |m 1331722098:Favaro, Luigi |d 130000 |d 130300 |e 130000PF1331722098 |e 130300PF1331722098 |k 0/130000/ |k 1/130000/130300/ |p 2 | ||
| 999 | |a KXP-PPN1890705632 |e 4534754752 | ||
| BIB | |a Y | ||
| SER | |a journal | ||
| JSO | |a {"person":[{"given":"Ranit","family":"Das","role":"aut","roleDisplay":"VerfasserIn","display":"Das, Ranit"},{"roleDisplay":"VerfasserIn","display":"Favaro, Luigi","role":"aut","family":"Favaro","given":"Luigi"},{"role":"aut","roleDisplay":"VerfasserIn","display":"Heimel, Theo","given":"Theo","family":"Heimel"},{"display":"Krause, Claudius","roleDisplay":"VerfasserIn","role":"aut","family":"Krause","given":"Claudius"},{"family":"Plehn","given":"Tilman","roleDisplay":"VerfasserIn","display":"Plehn, Tilman","role":"aut"},{"family":"Shih","given":"David","display":"Shih, David","roleDisplay":"VerfasserIn","role":"aut"}],"title":[{"title":"How to understand limitations of generative networks","title_sort":"How to understand limitations of generative networks"}],"note":["Veröffentlicht: 25. Januar 2024","Gesehen am 05.06.2024"],"type":{"bibl":"article-journal","media":"Online-Ressource"},"recId":"1890705632","language":["eng"],"name":{"displayForm":["Ranit Das, Luigi Favaro, Theo Heimel, Claudius Krause, Tilman Plehn and David Shih"]},"origin":[{"dateIssuedKey":"2024","dateIssuedDisp":"25 January 2024"}],"id":{"doi":["10.21468/SciPostPhys.16.1.031"],"eki":["1890705632"]},"physDesc":[{"extent":"32 S.","noteIll":"Illustrationen"}],"relHost":[{"physDesc":[{"extent":"Online-Ressource"}],"id":{"eki":["881391751"],"zdb":["2886659-9"],"issn":["2542-4653"]},"origin":[{"dateIssuedDisp":"[2016]-","publisher":"SciPost Foundation","publisherPlace":"Amsterdam"}],"language":["eng"],"recId":"881391751","note":["Gesehen am 06.03.17"],"disp":"How to understand limitations of generative networksSciPost physics","type":{"bibl":"periodical","media":"Online-Ressource"},"part":{"year":"2024","issue":"1","pages":"1-32","volume":"16","text":"16(2024), 1, Artikel-ID 031, Seite 1-32","extent":"32"},"pubHistory":["Vol. 1, issue 1 (September/October 2016)-"],"title":[{"title":"SciPost physics","title_sort":"SciPost physics"}]}]} | ||
| SRT | |a DASRANITFAHOWTOUNDER2520 | ||